Sleep disturbance is a disruptive symptom shared by the spectrum of progressive dementing illnesses, and its presence often precipitates decisions by families to seek institutional care for patients. Normal sleep-wake regulation is characterized by an oscillatory, circadian, alerting process and a linear, sleep-inducing process building need to sleep as a function of the duration of prior wakefulness. In our previous studies in this population, we have found diagnosis-specific circadian abnormalities in men, which implicate central, circadian dysfunction in the etiology of sleep-wake disturbance in Alzheimer's disease, frontotemporal degeneration, and Lewy body disease. In addition, we have found abnormalities in SCN cellular populations in men linked with specific circadian and behavioral changes in Alzheimer's disease. We now wish to build upon these initial findings and expand our studies to the extra-SCN circadian system as well as the SCN itself. We propose for this funding period to test four hypotheses. 1) Polysomnographic sleep in AD will be more disturbed in patients with large phase-delays of their circadian core-body temperature rhythm characterized by reduced sleep efficiency add longer sleep latency. Sleep will be more fragmented in patients with FTD compared to AD patients with an increased number of awakenings. 2) Female patients with probable AD will have similady delayed phase of temperature and activity as male patients and normal controls; 3) Noctumal agitation and restlessness, seen in AD, results from loss of serotonergic innervation of the suprachiasmatic nuclei and will be detectable as lower RIA serotonin transporter protein (5-HTT) in SCN compared to FTD patients and controls. In addition, measurements of nocturnal agitation will be higher in AD patients with lower 5-HTT; and 4) Patients with FTD wilt have lowered levels of orexin/hypocretin in target tissue of perifonical area of the hypothalamus, locus coeruleus, midline thalamus and/or dorsal raphe nuclei compared to AD and controls. The extent of dissociation of activity and temperature will be related to the 10ss of orexin/hypocretin in patients carrying the same dementia diagnoses. To accomplish these objectives we will study patients with progressive dementing illnesses collecting core-body temperature, polysomnographic and locomotor activity data every 6 months and followed by post-mortem neuropathological studies in addition to diagnosis-based neuropathological studies.